Advertisement

Anti-steganalysis for image on convolutional neural networks

  • Shiyu Li
  • Dengpan YeEmail author
  • Shunzhi Jiang
  • Changrui Liu
  • Xiaoguang Niu
  • Xiangyang Luo
Article
  • 64 Downloads

Abstract

Nowadays, convolutional neural network (CNN) based steganalysis methods achieved great performance. While those methods are also facing security problems. In this paper, we proposed an attack scheme aiming at CNN based steganalyzer including two different attack methods 1) the LSB-Jstego Gradient Based Attack; 2) LSB-Jstego Evolutionary Algorithms Based Attack. The experiment results show that the attack strategies could achieve 96.02% and 90.25% success ratio separately on the target CNN. The proposed attack scheme is an effective way to fool the CNN based steganalyzer and in addition demonstrates the vulnerability of the neural networks in steganalysis.

Keywords

Anti-steganalysis CNN Adversarial example 

Notes

Acknowledgements

An earlier version of this paper was presented at the 4th International Conference on Cloud Computing and Security, 8-10, June 2018, Haikou, China. This work was partially supported by the National Key Research Development Program of China (2016QY01W0200), the National Natural Science Foundation of China NSFC (U1636101,U1636219, U1736211).

References

  1. 1.
    Bas P, Filler T, Pevnỳ T (2011) Break our steganographic system: the ins and outs of organizing boss. In: International workshop on information hiding, Springer, pp 59–70Google Scholar
  2. 2.
    Das S, Suganthan PN (2011) Differential evolution: a survey of the state-of-the-art. IEEE Trans Evol Comput 15(1):4–31CrossRefGoogle Scholar
  3. 3.
    Floreano D, Mattiussi C (2008) Bio-inspired artificial intelligence: theories, methods, and technologies. MIT Press, CambridgeGoogle Scholar
  4. 4.
    Fridrich J, Kodovsky J (2012) Rich models for steganalysis of digital images. IEEE Trans Inf Forensics Secur 7(3):868–882CrossRefGoogle Scholar
  5. 5.
    Goodfellow IJ, Shlens J, Szegedy C (2014) Explaining and harnessing adversarial examples. arXiv:14126572
  6. 6.
    Krizhevsky A, Hinton G (2009) Learning multiple layers of features from tiny images. Tech. rep., CiteseerGoogle Scholar
  7. 7.
    Kurak C, McHugh J (1992) A cautionary note on image downgrading. In: Proceedings 8th annual computer security applications conference, 1992. IEEE, pp 153–159Google Scholar
  8. 8.
    Kurakin A, Goodfellow I, Bengio S (2016) Adversarial examples in the physical world. arXiv:160702533
  9. 9.
    LeCun Y (1998) The mnist database of handwritten digits. http://yannlecuncom/exdb/mnist/
  10. 10.
    Luo X, Song X, Li X, Zhang W, Lu J, Yang C, Liu F (2016) Steganalysis of hugo steganography based on parameter recognition of syndrome-trellis-codes. Multimed Tools Appl 75(21):13557–13583CrossRefGoogle Scholar
  11. 11.
    Ma Y, Luo X, Li X, Bao Z, Zhang Y (2018) Selection of rich model steganalysis features based on decision rough set α-positive region reduction. IEEE Trans Circuits Syst Video Technol  https://doi.org/10.1109/TCSVT.2018.2799243
  12. 12.
    Moosavi-Dezfooli SM, Fawzi A, Frossard P (2016) Deepfool: a simple and accurate method to fool deep neural networks. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 2574–2582Google Scholar
  13. 13.
    Moosavi-Dezfooli SM, Fawzi A, Fawzi O, Frossard P (2017) Universal adversarial perturbations. arXiv preprintGoogle Scholar
  14. 14.
    Nguyen A, Yosinski J, Clune J (2015) Deep neural networks are easily fooled: high confidence predictions for unrecognizable images. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 427–436Google Scholar
  15. 15.
    Pevny T (2011) Detecting messages of unknown length. In: Media watermarking, security, and forensics III, international society for optics and photonics, vol 7880, p 78800TGoogle Scholar
  16. 16.
    Pibre L, Pasquet J, Ienco D, Chaumont M (2016) Deep learning is a good steganalysis tool when embedding key is reused for different images, even if there is a cover sourcemismatch. Electron Imaging 2016(8):1–11CrossRefGoogle Scholar
  17. 17.
    Qian Y, Dong J, Wang W, Tan T (2015) Deep learning for steganalysis via convolutional neural networks. In: Media watermarking, security, and forensics 2015, international society for optics and photonics, vol 9409, p 94090JGoogle Scholar
  18. 18.
    Storn R, Price K (1997) Differential evolution–a simple and efficient heuristic for global optimization over continuous spaces. J Glob Optim 11(4):341–359MathSciNetCrossRefGoogle Scholar
  19. 19.
    Su J, Vargas DV, Kouichi S (2017) One pixel attack for fooling deep neural networks. arXiv:171008864
  20. 20.
    Szegedy C, Zaremba W, Sutskever I, Bruna J, Erhan D, Goodfellow I, Fergus R (2013) Intriguing properties of neural networks. arXiv:13126199
  21. 21.
    Upham D (1997) Jsteg. Software available at ftp funet fiGoogle Scholar
  22. 22.
    Wang J, Li T, Shi YQ, Lian S, Ye J (2017) Forensics feature analysis in quaternion wavelet domain for distinguishing photographic images and computer graphics. Multimed Tools Appl 76(22):23721–23737CrossRefGoogle Scholar
  23. 23.
    Winkler S, Mohandas P (2008) The evolution of video quality measurement: from psnr to hybrid metrics. IEEE Trans Broadcast 54(3):660–668CrossRefGoogle Scholar
  24. 24.
    Wu S, Zhong SH, Liu Y (2017) A novel convolutional neural network for image steganalysis with shared normalization. arXiv:171107306
  25. 25.
    Xu G, Wu HZ, Shi YQ (2016) Structural design of convolutional neural networks for steganalysis. IEEE Signal Process Lett 23(5):708–712CrossRefGoogle Scholar
  26. 26.
    Ye J, Ni J, Yi Y (2017) Deep learning hierarchical representations for image steganalysis. IEEE Trans Inf Forensics Secur 12(11):2545–2557CrossRefGoogle Scholar
  27. 27.
    Zhang Y, Qin C, Zhang W, Liu f, Luo X (2018) On the fault-tolerant performance for a class of robust image steganography. Signal Process 146:99–111CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2018

Authors and Affiliations

  • Shiyu Li
    • 1
  • Dengpan Ye
    • 1
    Email author
  • Shunzhi Jiang
    • 1
  • Changrui Liu
    • 1
  • Xiaoguang Niu
    • 2
  • Xiangyang Luo
    • 3
  1. 1.School of Cyber Science and EngineeringWuhan UniversityWuhanChina
  2. 2.School of Computer ScienceWuhan UniversityWuhanChina
  3. 3.State Key Laboratory of Mathematical Engineering and Advanced ComputingZhengzhouChina

Personalised recommendations